• http://rud.is/b Less infosec, more R & vis
• http://twitter.com/hrbrmstr
• http://github.com/hrbrmstr
• bob@rudis.net (if you like waiting for responses)

head(pressure, 3)

##   temperature pressure
## 1           0   0.0002
## 2          20   0.0012
## 3          40   0.0060

summary(pressure)

##   temperature     pressure       
##  Min.   :  0   Min.   :  0.0002  
##  1st Qu.: 90   1st Qu.:  0.1800  
##  Median :180   Median :  8.8000  
##  Mean   :180   Mean   :124.3367  
##  3rd Qu.:270   3rd Qu.:126.5000  
##  Max.   :360   Max.   :806.0000

z <- seq(-10, 10, 0.01)
scatterplot3js(cos(z), sin(z), z, color=rainbow(length(z)))

forceNetwork(Links = MisLinks, Nodes = MisNodes, Source = "source",
             Target = "target", Value = "value", NodeID = "name",
             Group = "group", opacity = 0.4)

dest <- factor(sprintf("%.2f:%.2f",flights[,3], flights[,4]))
freq <- sort(table(dest), decreasing=TRUE)
frequent_destinations <- names(freq)[1:10]
idx <- dest %in% frequent_destinations
frequent_flights <- flights[idx, ]
latlong <- unique(frequent_flights[,3:4])
earth <- system.file("images/world.jpg",  package="threejs")
globejs(img=earth, lat=latlong[,1], long=latlong[,2], 
        arcs=frequent_flights, arcsHeight=0.3, arcsLwd=2, 
        arcsColor="#ffff00", arcsOpacity=0.15, atmosphere=TRUE)

• Dynamic (like JavasScript & Python)
• Has variables (like JavasScript & Python)
• …functions (like JavaScript & Python)
• …loops (like JavaScript & Python)
• …and, help from friends (packages) (like Node or Python modules)

It's "vectorized" (think map() or [ for ])

a <- 1:10
sum(a)

## [1] 55

It really likes something called "data frames" (Python does too, now)

head(iris)

##   Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 1          5.1         3.5          1.4         0.2  setosa
## 2          4.9         3.0          1.4         0.2  setosa
## 3          4.7         3.2          1.3         0.2  setosa
## 4          4.6         3.1          1.5         0.2  setosa
## 5          5.0         3.6          1.4         0.2  setosa
## 6          5.4         3.9          1.7         0.4  setosa

It has affintity for arcane punctuation:

`huh?` <- iris$Sepal.Length[[2]] * 3 %>% sqrt()
print(`huh?`)

## [1] 8.487049

And, complex+efficient algorithms can be confusing:

dat <- readLines(textConnection(" 3   weeks,   2  days,  4 hours
 4 week,  6 days,  12 hours 
4 day, 3 hours
 7 hours
8  hour"))

sapply(str_split(str_trim(dat), ",[ ]*"), function(x) {
  sum(sapply(x, function(y) {
    bits <- str_split(str_trim(y), "[ ]+")[[1]]
    duration(as.numeric(bits[1]), bits[2])
  })) / 3600
})

## [1] 556 828  99   7   8

library(ggplot2)
g1 <- ggplot(mtcars, aes(x=wt, y=mpg)) + geom_smooth() + geom_point()
print(g1)

This is all it takes to turn that plot into an editable/usable SVG graphic:

ggsave(g1, "img/g1.svg")

<?xml version="1.0" encoding="UTF-8"?>
<svg xmlns="http://www.w3.org/2000/svg" 
     xmlns:xlink="http://www.w3.org/1999/xlink"
    width="819pt" height="425pt" 
    viewBox="0 0 819 425" version="1.1">
<defs>
<g>
<symbol overflow="visible" id="glyph0-0">
<path style="stroke:none;" 
      d="M 0.3125 0 L 0.3125 -6.875 L 5.765625 -6.875 L 5.765625 0 Z M 4.90625 -0.859375 L 4.90625 -6.015625 L 1.171875 -6.015625 L 1.171875 -0.859375 Z "/>
</symbol>
<symbol overflow="visible" id="glyph0-1">
<path style="stroke:none;" 
      d="M 0.921875 -4.75 L 0.921875 -5.390625 C 1.523438 -5.453125 1.945312 -5.550781 2.1875 -5.6875 C 2.425781 -5.832031 2.609375 -6.164062 2.734375 -6.6875 L 3.390625 -6.6875 L 3.390625 0 L 2.5 0 L 2.5 -4.75 Z "/>

library(ggthemes)
g1 + theme_economist()

library(xkcd)
g1 + theme_xkcd()

g1 + theme_excel()

dat <- read.csv("data/download.csv", skip=4, header=TRUE, stringsAsFactors=FALSE)
dat <- head(dat, -2)
head(dat)

##   GeoFips                                                     GeoName
## 1   00999                     United States (Nonmetropolitan Portion)
## 2   10180                 Abilene, TX (Metropolitan Statistical Area)
## 3   10420                   Akron, OH (Metropolitan Statistical Area)
## 4   10500                  Albany, GA (Metropolitan Statistical Area)
## 5   10540                  Albany, OR (Metropolitan Statistical Area)
## 6   10580 Albany-Schenectady-Troy, NY (Metropolitan Statistical Area)
##   X2013
## 1  87.7
## 2  91.3
## 3  88.9
## 4  84.3
## 5  93.5
## 6  99.0

dat$GeoName <- gsub(" \\(Metropolitan Statistical Area\\)", "", 
                    dat$GeoName)
dat$GeoFips <- sprintf("%05d", as.numeric(dat$GeoFips))
head(dat)

##   GeoFips                                 GeoName X2013
## 1   00999 United States (Nonmetropolitan Portion)  87.7
## 2   10180                             Abilene, TX  91.3
## 3   10420                               Akron, OH  88.9
## 4   10500                              Albany, GA  84.3
## 5   10540                              Albany, OR  93.5
## 6   10580             Albany-Schenectady-Troy, NY  99.0

library(jsonlite)
dat <- readJSON("that horrible URL")

dat <- dat$BEAAPI$Results$Data
dat$X2013 <- as.numeric(dat$DataValue)
dat$GeoName <- gsub(" \\(Metropolitan Statistical Area\\)", "", 
                    dat$GeoName)
dat$GeoFips <- sprintf("%05d", as.numeric(dat$GeoFips))
head(dat[,c(1,2,8)])

##   GeoFips                     GeoName X2013
## 1   10180                 Abilene, TX  91.3
## 2   10420                   Akron, OH  88.9
## 3   10500                  Albany, GA  84.3
## 4   10540                  Albany, OR  93.5
## 5   10580 Albany-Schenectady-Troy, NY  99.0
## 6   10740             Albuquerque, NM  97.1

library(httr)
response <- GET("http://bea.gov/api/data/",
                query=list(
                  UserID=Sys.getenv("BEA_API_TOKEN"),
                  method="GetData",
                  datasetname="RegionalData",
                  KeyCode="RPPALL_MI",
                  Year="2013",
                  ResultFormat="json"
                ))
dat <- fromJSON(content(response, as="text"))

Same cleanup as we did in the raw URL version

dat <- dat$BEAAPI$Results$Data
dat$X2013 <- as.numeric(dat$DataValue)
dat$GeoName <- gsub(" \\(Metropolitan Statistical Area\\)", "", 
                    dat$GeoName)
dat$GeoFips <- sprintf("%05d", as.numeric(dat$GeoFips))
head(dat[,c(1,2,8)])

##   GeoFips                     GeoName X2013
## 1   10180                 Abilene, TX  91.3
## 2   10420                   Akron, OH  88.9
## 3   10500                  Albany, GA  84.3
## 4   10540                  Albany, OR  93.5
## 5   10580 Albany-Schenectady-Troy, NY  99.0
## 6   10740             Albuquerque, NM  97.1

library(noncensus)
data(counties)
xlate <- data.frame(fipscounty=sprintf("%s%s", counties$state_fips, 
                                       counties$county_fips), 
                    cbsa=counties$CBSA,
                    stringsAsFactors=FALSE)
dat <- merge(dat[,c(1,2,8)], xlate[,c("cbsa", "fipscounty")], 
             by.x="GeoFips", by.y="cbsa")
head(dat)

##   GeoFips     GeoName X2013 fipscounty
## 1   10180 Abilene, TX  91.3      48441
## 2   10180 Abilene, TX  91.3      48253
## 3   10180 Abilene, TX  91.3      48059
## 4   10420   Akron, OH  88.9      39133
## 5   10420   Akron, OH  88.9      39153
## 6   10500  Albany, GA  84.3      13177

• Display a US map county choropleth with the counties filled according the RPP value
• The counties are not all represented and we don't need a billion small polygons left over so we'll outline the states for context
• We'd like to add contextual information via popup to show the discrete data value and the name of the metro area
• A legend would be good

• Raw JS or jQuery + CSS
• Straight D3
• Kartograph
• Datamaps
• Leaflet [Ex: http://leafletjs.com/examples/choropleth-example.html] (Perfect data. ~170 lines)

Get map data

library(rgdal)
URL <- "http://bl.ocks.org/mbostock/raw/4090846/us.json"
fil <- basename(URL)
if (!file.exists(fil)) download.file(URL, fil)

# read state borders from the file
states <- readOGR(fil, "states", stringsAsFactors=FALSE, 
                  verbose=FALSE)

# read county borders from the file
county <- readOGR(fil, "counties", stringsAsFactors=FALSE, 
                  verbose=FALSE)

We don't want to display all the counties, so we'll subtract out the ones that aren't in our data set.

rpp_counties <- subset(county, id %in% dat$fipscounty)
rpp_counties <- merge(rpp_counties, dat, 
                      by.x="id", 
                      by.y="fipscounty", 
                      all.x=TRUE)

We need to setup the color scale (really similar to how you'd do it in JS)

library(leaflet)
pal <- colorBin("BrBG", range(rpp_counties$X2013), bins=5)
rpp_counties$color <- pal(rpp_counties$X2013)

leaflet() %>%
  addProviderTiles("Acetate.terrain") %>%
  addPolygons(data=rpp_counties, weight=0.25,
              fillColor=~color, color="black", fillOpacity=1,
              popup=~sprintf("In %s, <span style='font-weight:700'>%s</span> has the purchasing power of $100.00.",
                             htmlEscape(GeoName),
                             htmlEscape(dollar(X2013)))) %>%
  addPolygons(data=states, weight=0.5, fillColor="white", fillOpacity=0, fill=FALSE, color="#525252") %>%
  addLegend(position="bottomright", pal=pal, values=rpp_counties$X2013, labFormat=labelFormat("$"), opacity = 1) %>%
  setView(-74.0059, 40.7127, 6)

General

• built-in support for CSV/TSV/general delimited & fixed-width
• readr / rio faster & more robust compatibility
• readxl (and others) for raw Excel reading
• googlesheets
• data.table (large data)
• numerous packages to read statistical data files

Web Scraping / API

• httr (like curl command line but better)
• rvest (more structured web page scraping)
• jsonlite (JSON)
• XML / xml2 (XML)
• Rselenium (headless browser & DOM scraping)
• V8 (the V8 engine in R)

Database

• dplyr
• RPostgreSQL
• RMySQL
• rredis
• mongolite
• RSQLite

Database

• dplr
• RPostgreSQL
• RMySQL
• rredis
• mongolite
• RSQLite

• Almost every useful public API covered
• Virtually every "big data" store including AWS/S3
• If something is missing, complain on Twitter and there'll be a package in a week

• files (CSV/JSON/XML)
• database write
• S3 upload
• API "put"

• plumber http://plumber.trestletech.com/ (think "Flask")
• httpuv https://github.com/rstudio/httpuv (basic web & websocket server)

• htmlwidgets http://www.htmlwidgets.org/
• The widget gallery http://hafen.github.io/htmlwidgetsgallery/ (i've got 3! :-)
• You've already seen one! (leaflet)

library(taucharts)
data(cars_data)
tauchart(cars_data) %>%
  tau_point("milespergallon", c("class", "price"), color="class") %>%
  tau_trendline() %>% tau_legend()

• devtools::create("/path/to/new/package")
• setwd("/path/to/new/package")

or use RStudio

• htmlwidgets::scaffoldWidget()

• devtools::build()
• devtolls::install()

or use RStudio